At present, artificial intelligence (AI) has been widely used in the diagnosis and treatment of various ophthalmological diseases, but there are still many problems. Due to the lack of standardized test sets, gold standards, and recognized evaluation systems for the accuracy of AI products, it is difficult to compare the results of multiple studies. When it comes to the field of image generation, we hardly have an efficient approach to evaluating research results. In clinical practice, ophthalmological AI research is often out of touch with actual clinical needs. The requirements for the quality and quantity of clinical data put more burden on AI research, limiting the transformation of AI studies. The prediction of systemic diseases based on fundus images is making progressive advancement. However, the lack of interpretability of the research lower the acceptance. Ophthalmology AI research also suffer from ethical controversy due to unconstructed regulations and regulatory mechanisms, concerns on patients’ privacy and data security, and the risk of aggravating the unfairness of medical resources.
Objective To explore the feasibility and effect of remote medical education model using online film reading training to improve the ability of ophthalmologists in the Xinjiang Uygur Autonomous Region (hereinafter referred to as “Xinjiang Region”) in diagnosing fundus diseases. MethodsThe three-level film reading training system of Xinjiang Production and Construction Corps system division hospital-Corps Hospital-Peking Union Medical College Hospital was established. From June 2022 to January 2023, 4 159 posterior color fundus images were continuously collected from Department of Ophthalmology of Xinjiang Corps Hospital and 4 divisional hospitals in the Corps medical system. Among them, hypertensive retinopathy, diabetic retinopathy, exudative age-related macular degeneration (AMD), atrophic AMD and retinal vein occlusion were 3 073, 651, 43, 186 and 206 cases, respectively. The images were divided into 3 rounds (first, second and last) according to the proportion of diseases. The doctors who participated in the training (hereinafter referred to as the "training") were 15 ophthalmologists from the Corps Hospital of Xinjiang Region and the division hospital of the Corps system. There were 7 male and 8 female. Age was (38.1±4.0) years. The titles of senior, deputy senior, intermediate and junior are 1, 6, 5 and 3 respectively; Bachelor's degree and master's degree are 13 and 2 respectively. The working time of fundus disease specialty was (9.6±3.3) years. The film reading system training was conducted before the first round of labeling, and after each round of film reading, the doctors of Peking Union Medical College Hospital gave feedback and explanation on the film reading results. The diagnostic consistency, sensitivity and specificity were compared by paired sample t test. Spearman or Pearson correlation analysis was conducted between the improvement of diagnostic level and professional title, education, age and working hours of ocular fundus disease. ResultsAll the participating doctors completed the first, second and last reading. After each round of film reading, the film reading summary was carried out for 2 hours. The average diagnostic agreement rates of participating physicians were 53.0%, 67.0% and 75.0%, respectively. The sensitivity and specificity were 0.38, 0.69, 054 and 0.66, 0.85, 0.96, respectively. There was significant difference between the first and last examination (P＜0.001). The sensitivity of the second reading was significantly higher than that of the first reading, and the sensitivity of the last reading was significantly lower than that of the second reading, with statistical significance (P＜0.05). The specificity of the second reading was significantly higher than that of the first reading, and the last reading was significantly higher than that of the second reading, with statistical significance (P＜0.05). There was no significant correlation (P＞0.05) between the improvement of diagnostic level of participating physicians and educational background (Rho=0.07), professional title (Rho=0.13), age (r=0.20), and working time of ophthalmofundus disease specialty (r=0.26).ConclusionsRelying on the three-level online telemedicine training, it can improve the ability of ophthalmologists in Xinjiang region to diagnose fundus diseases. The preliminary telemedicine education model has demonstrated potential for feasibility and effectiveness in remote areas with inadequate medical resources.